3.4.5  Amending the methodology of a review

In addition to searching for new studies and revising the review question or study selection criteria, maintenance of a review may include amendment of the methodology of the review (Shea 2006). Methodological advances in systematic review conduct since publication of the original review may result in a need to revise or extend the methods of a review during an update. Review authors may decide to include a new analysis strategy in their updated review (for example, using statistical methods not previously available in RevMan). The introduction of ‘Risk of bias’ (Chapter 8) and ‘Summary of findings’ (Chapter 11) tables with RevMan 5, while not mandatory, provides the opportunity for reviews to be updated to include these new methods. Where a ‘Risk of bias’ table is to be added to a review, authors should decide whether to revisit the critical appraisal of studies included in previous versions of the review, updating all assessments of risk of bias, or whether to apply these new methods only to studies added in the update. In the published version of the review, a ‘Risk of bias’ table should be generated including only those studies where data are entered (i.e. without blank rows).

 

As part of a review update, authors may wish to include a ‘Summary of findings’ table (Chapter 11). Outcomes selected for presentation in the ‘Summary of findings’ table should be those of importance to people making decisions about health care (usually the primary outcomes of the review), and should be selected prior to commencement of the update to reduce the risk of selectively reporting outcomes with significant results rather than those of importance.

 

Changes to methodology may imply changes to the original protocol of the review. These changes, and their justifications, must be explicitly provided in the ‘Differences between protocol and review’ section and the ‘What’s new’ table.